The accuracy claim is not exactly true. The best clocks are so astonishingly good only for measurements of shorter times, comparable to human activities: That's all we need, of course, but let's get the claim right. The issue is red noise; The most familiar example is flicker noise that rises as 1/f - louder at low frequencies, very much louder at very low frequencies that become important for measurements made over very long times. Here is the precise statement: When measuring an interval a second long, the best modern clocks make an error in that measurement, compared to a second, that is smaller than the second, compared to the age of the universe. Much less punchy. No puzzle why you usually hear the simpler, wrong, version.. NIST has even given up and blessed the simpler one.

It takes three clocks. You run the three clocks at the same time, recording the differences of the three pairs - offset frequency and noise in that frequency - to gather data. You do the "three cornered hat" which processes the data and attributes to each clock its share of the deviations from some ideal. Then you take that ideal as your best estimate of the frequency of the atomic transition. Precise timekeeping goes on a long way further than that, into spectral densities and time scales based on a larger clock ensemble, but that's the first step..

Let's hear it for stability of one's working environment! Just this morning I explained to a software sales person that I do not use "software as a service" or any software that requires recurrent contact with a server for continued function. For instance, I quit updating my copy of MathCAD when the next version would have required recurrent server contact. I pay for maintenance when it is available, but I need to be able to open a 20-year-old file in its native application even when the vendor of that application has abandoned it. And yes, I have XP up in a virtual machine. As you might guess, there is some history behind that policy.

C'mon, guys, it's radiation pressure; all conventional physics and it really can work. A photon of energy hv carries momentum hv/c where c is the speed of light. A beam of EM radiation of power W exerts a reaction force of W/c on the object that emits it. Each photon is a bit of energy, which is equivalent to a bit of mass as Einstein taught us. So yes, something is being ejected - photons. And yes the craft gets lighter as they leave. The craft really is using up reaction mass, and energy to accelerate it. It really does work for the same reason that a conventional rocket works. The only weird thing is that the exhaust velocity is the speed of light, so the specific impulse of the "fuel" is about 30 million seconds, not a few hundred as with chemical fuels. So the craft gets lighter very slowly. This is the natural engine for a nuclear powered long-mission spacecraft. It is hard to make much thrust this way, but you don't use up your fuel very fast. This electromagnetic thrust is well known to guys designing solar power satellites; the expected thrust from 10GW of power beam is 33 Newtons. Sunlight being absorbed by the solar panels exerts an even larger force. Another effect of light pressure is the force on a solar sail. (A so-called radiometer spinner is a quite different effect that you can look up.) Remember the spacecraft (Voyager?) that was experiencing anomalous acceleration in the outer solar system when the expected accelerations were extremely small? A thermal power generator was radiating its waste head in mostly one direction. Add in the tiny thrust from that and it all came out even. So this effect has even been tested in space, although unintentionally.

A Patent can be issued for a novel and useful composition of matter or process. A theorem is neither, and is indeed not patentable. But an automated process, like the bottle capper at the end of a beer bottling production line, is carrying out a potentially patentable process, even though it may be implemented in a way that depends crucially on the operation of an embedded control program. Agreed so far? Let's take another step: how about the PageRank algorithm that processes information in a way that grew up in an empirically-driven manner without any central core of mathematics? I think that there is a sensible argument to be made that we are still in patentable territory under present law. Dammit. I'd like to see most software patents eliminated, but their continued existence is not good evidence against SCOTUS understanding of software.

Curious indeed! If you are loading the memory image from the same linker output file format as the debugger uses, for instance if you are loading with the debugger itself, then I would not expect the problem to occur. If you are really loading with one file format and verifying from another, and still have no trouble, please post that fact. It would be good to know that EW for ARM is better than EW for MSP430 in that way. The simplest way to account for my observations is to assume that there is an entirely separate linker for each output format at least for the MSP430, and that the code that chooses link order is not always the same from one to another. The way standards were set and enforced during development could then be very different for the two flavors, and indeed the algorithm might be always the same for the ARM flavor. I am about to decide what tools to use for an ARM project. I like most things about IAR, and if you can verify the memory image successfully using different linker output formats for original load and for the debugger when verifying I would almost certainly use IAR on ARM instead of something else.

I write embedded control firmware for MSP430 processors, building and debugging with IAR Embedded Workbench. In production I build each version to two targets with identical source files but with the single change of different loader output file formats, one for the TI gang programmer used in production, another for the field update loader that we must sometimes distribute to update customers' systems. A third output format (with debug information) is needed if I am going to go in through the JTAG port to do any debugging. Surprise: the resulting memory images from any two of these builds using the same source files have not been identical any time that I have checked. There is no hash nor any date field by the time the image is loaded and I make the comparison with the contents of target hardware memory. In this case, the linker does not always place modules in the same order, and that seems to account for the difference. As far as I can tell they are always linked correctly and so far the program images always seem to have identical functionality, but it means that I cannot use the memory compare function of the JTAG debugger to verify a memory image that was loaded with either the Gang programmer or our field update loader. I asked IAR about this, and they said that yes, the module order was not guaranteed to be consistent between loader output file formats. So I can be sure that each of these build output files does correspond to a known source, and the same source, and all of them work if any of them do, but the memory images they produce fail comparison. Grumble, grumble.

That fallacy is a caricature, of course. With Copernicus, again look at the details. Copernicus got away with the science per se. What got him in trouble was defying church authority about how to phrase it. It didn't help at all that he put the pope's arguments in the mouth of his fool character in the dialogues. The Catholic church has a genuinely bad record about lots of things, but it is, for instance, one of the few to acknowledge plainly and officially today that evolution is genuine science to be taken seriously. It really is true that a large fraction of the early work in any science was done by religious people, who in a paraphrase of Augustine's words felt that they were "reading that other book written by God: nature". And also true that the idea that the world was governed by a single orderly scheme, which arose strongly in the monotheistic religions, was an important precursor of science. Fools there are, and fools there have always been. Most of the fools are religious whenever and wherever most people are religious. There is a genuine tension between the approaches taken by science and religion, of course, and the competition for loudest modern anti-science fools is between the religious right and the economic interests that urge them on. But the idea that the two must be eternal enemies is a new one and one I don't think is correct. The names I didn't have with me when writing an earlier post were John William Draper and Andrew Dixon White. They are largely responsible for the impression widely held today that science and religion must always be enemies. That was not a widely held idea before late in late in the 19th century. For this history, see p139 and the following section of Understanding Fundamentailsm and Evangelicalism, by George M Marsden ISBN 978-0-8028-0539-3. The Great Courses videos on "Science and Religion" offer much the same story. So what is a self-described "non-religious" person doing with all these references? I'm interested in religion as a phenomenon, and being a scientist I'm in the habit of looking at the evidence.

I had intelligent, devout parents and grew up in a conservative religious backwater. Our pastor was a nominally Lutheran biblical literalist. I slowly pulled myself loose from the science denial of my church, and went on to become a scientist myself (Physics). My path would have been easier had I known then about Augustin and his kin who a millennium or more ago also had to pull themselves away from simplistic interpretations of the Bible. I ended up not religious myself, but I can respect my friends, including scientists, who are religious. The frightened religious conservatives we see so commonly in the US today are not representative of the best in the world's religious traditions, nor the best in Christianity, and they are not even typical of thoughtful Christians that we can see in a broad historical view. The supposed eternal conflict between science and religion is a late-developing meme, propagated in the late 18th century by a couple of folks (I do not have the reference here with me) for their own purposes as part of the professionalization of science, which had previously been an amateur's realm. im-thatoneguy may have had a bad early experience with Christians, as did the most virulently anti-christian of my friends, but he should keep in mind that the loudest Christians we hear today in the US are a recent anomaly, and are a caricature of Christianity. We need to look a bit deeper to see the real relation between science and religion, and our guest for the last two days has kindly pointed us into that deeper realm. I thank him for it, and I think that we all should do that.

See http://www.rdmag.com/RD100-Awards-Rounding-The-Edges-On-Superconductor-Wires/ which I reached via the sapphire outfit's site. The sapphire is a substrate for epitaxial deposition of an unspecified superconductor. It is not the conductor and the story is making more sense now.

Someone will put up a trading program that detects these flash crashes and profits from them.